Background: A nursing doctoral candidate at a private university in Kerala approached Thesis Writing Cafe in October 2024, eight months behind schedule on her data analysis chapter. Her research examined nurse burnout predictors across three hospital settings using a validated 52-item questionnaire administered to 287 respondents.
The challenge: The researcher had collected her data but had no prior SPSS experience. Her supervisor had flagged that the descriptive statistics were incomplete, the reliability analysis (Cronbach alpha) had not been conducted per instrument subscale, and the inferential statistics plan logistic regression with three predictors had not accounted for multicollinearity or missing data imputation.
What we did: Dr. Krishnamurthy reviewed the raw SPSS dataset and conducted a full audit of the existing analysis. The work involved: (1) recoding and cleaning 14 reverse-scored items the researcher had left uncorrected; (2) running subscale-level reliability analysis all four subscales returned Cronbach alpha 0.79; (3) replacing listwise deletion with multiple imputation for 11 missing cases (3.8%); (4) checking the assumption of multicollinearity via VIF scores all below 3.2; and (5) running binary logistic regression with odds ratios and 95% confidence intervals reported per APA 7th edition.
Outcome: The revised data analysis chapter was submitted to the supervisor within three weeks. Minor revisions were requested on one table footnote formatting. The chapter was approved in full at the second submission. The researcher subsequently engaged us for her results interpretation and discussion chapter.
This case study is based on a real engagement. Identifying details have been anonymised in line with our confidentiality policy.